Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit


Your coursework must be submitted electronically via Blackboard by 3pm on Friday March 25th.   Any work handed in after this time will be subject to the following penalties: 10% of your marks lost per working day up to 5 working days. Do not write your names anywhere on your work, as marking will be anonymous.  Your student IDs should be included in the filenames but not your name; see further instructions on file naming and labelling in Section 3 below.  An extension, for bona fide reasons, may be allowed by prior agreement, but only well before the deadline; you can contact the Student Office if you would like to apply for an extension.  Computer crashes or file losses a day or two before the deadline will not be an acceptable reason for an extension. It is therefore advisable to keep back-up copies of your work. Components of the  project  will  receive  different  weightings  in  producing  your  final  mark:   40 marks for the exponential smoothing part, 20 for ARIMA, 20 for regression, 10 for the presentation slide, and 10 marks for the overall organization of your submitted material, including the description of your codes/files.

You are expected to complete the assessment in groups of 3 students; working alone or in a group of 2 could be accepted if there is a valid reason to do so. Please email the lecturers as soon as possible and no later than 2 weeks before the submission deadline if you are not able to form a group of 3 to complete the assignment.  All students in each group will get the same mark for their work and for any late submission, all the group will endure the same level of penalties as indicated above.

1. Background and analysis

In light of the recent United Nations Climate Change Conference that took place in Glasgow, Scotland, from 31 October to 13 November 2021, the UK government through its new Clean Green Initiative has employed you as a consultant.  Your task is to forecast the behaviour of a number of key environmental indicators until December  2022,  to help support the decision process for new policies to support the country’s efforts to reduce the impact of climate change.  The data is provided by a number of public organizations, including the Meteorological (Met) Office and the Office for National Statistics (ONS).

1.1. How to get the data. From the four weblinks given below, download the data sets and save them in xlsx or xls format.  The resulting files might have multiple columns or sheets; follow the corresponding instructions to access the data necessary for your analysis. Copy the data sets from the required columns as described below; i.e., MSTA, CH4, GMAF, and ET12, scrolling down, where necessary, to find the monthly observations.

(A) Global Mean Surface Temperature Anomaly (MSTA) in C:


MSTA: to get the data, see first table monthly box in the Global row; select the CSV-file type – the data is located in the Anomaly column.

(Source: The Meteorological Office, abbreviated as the Met Office, which is the United King- dom’s national weather service).

(B) Global Monthly Atmospheric Carbon Dioxide Levels (CH4):

https://gml.noaa.gov/webdata/ccgg/trends/ch4/ch4 mm gl.txt

CH4: see average monthly values in 4th column.

(Source:   Global Monitoring Laboratory of the USA National  Oceanic and Atmospheric Administration, an American scientific and regulatory agency within the United States De- partment of Commerce).                                                                                                           It is recommended that for time series in (B), you copy the data into text files, using, for example, Notepad, and then open the text files using Excel, as space delimited.  The files can then be saved as Excel workbooks.

(C) International Passenger Survey, UK visits abroad (GMAF):

https://www.ons.gov.uk/peoplepopulationandcommunity/leisureandtourism/datasets/interna tionalpassengersurveytimeseriesspreadsheet

GMAF: select the xlsx file; see data in GMAF column, scrolling down to the monthly data.

(Source: UK Office for National Statistics).                                                                             (D) UK inland monthly energy consumption (ET12), million tonnes of oil equivalent–xls file can be downloaded by clicking on the corresponding expression with the this link:


ET12: use the data in the Total unadjusted column of the Month worksheet. (Source: Department for Business, Energy & Industrial Strategy).

1.2. Tasks. As it so often happens in the real world, the data sets are of different lengths. You will have to use your own judgment in inspecting and preparing the data before carrying out any technical analysis. The analysis is in three parts:

(a) You are asked to take all four series separately and to forecast monthly behaviour until December 2022, using exponential smoothing-type forecasting methods.

(b) The Clean Green Initiative team have been satisfied in the past with exponential smoothing- type forecasting methods and are happy to see these techniques used in the analysis. How-   ever, they are interested in the possible use of the ARIMA methodology to predict MSTA.   You are asked to fit the ARIMA model to MSTA, for analysis in which you compare the use   of ARIMA forecasting and an exponential smoothing method.  You should make a recom-   mendation as to future use of ARIMA on this time series.

(c) The Clean Green Initiative team is interested to know whether global temperatures (that is, series MSTA) are affected by carbon dioxide levels, international air travel, and the con- sumption of fuels (as exemplified by series CH4, GMAF, and ET12).  Develop a multiple regression model, use it for prediction of MSTA until December 2022, and report on whether you think the model is satisfactory or not.

2. What you must produce

You must produce a technical report describing all the analysis done to select the most suit- able forecasting methods, as well as the results obtained. The report must be accompanied by a single-page slide summarizing your main results, and also the codes used to perform the technical analysis, as well as the resulting graphs.  More details on each of the aspects of the work are given in the next subsections.

2.1. The technical report. The technical report must follow the structure described in Subsection 2.5.  It should address the three parts of the analysis:  exponential smoothing, ARIMA, and regression. For each part, give details of the preliminary analysis, data prepa- ration, models chosen and analysis carried out. Also describe why each model was built and explain the analysis carried out, including an evaluation of the effectiveness of the models.

2.2. Single presentation slide. The executive board members of the Clean Green Initia- tive are particularly interested in knowing how the three methods (exponential smoothing, ARIMA, and regression) perform on the four variables mentioned above; i.e., MSTA, CH4, GMAF, and ET12.  You are asked to produce a single-page slide summarizing the main results of your analysis, in order to enable them to quickly grasp the results without neces- sarily having to read your technical report.  Where necessary, attention should be given to the comparison of the performance of the methods, while highlighting the best results. This slide will be judged on the suitability of its presentational style, clarity, and quality.

2.3. Python codes. You must also prepare and submit python codes that you use to gen- erate the results that will be included in your technical report. If any preliminary operations on your data are needed before applying/developing a python code for your analysis, it is fine to include this in the corresponding excel file containing your data sets.  However, you must complete all the main tasks of your analysis using python. You can use the codes from the course, use different ones or develop your own. Marking on this aspect of your work will not be based on how well you can program in python, but rather on the functionality of your codes and their relevance in the corresponding analysis.

To help us easily know what you do in each code, you must produce a single page document, as Appendix A to your technical report, to give a brief one or two sentences description of what it does. If you do any preliminary operations on your data in the excel file containing your data set, a line or two should also be included to describe this.

2.4. Analysis and forecast graphs. You are expected to produce graphs to illustrate your analysis in the technical report. Do not include these graphs in the main part of the report (Sections 1 - 3; see details in next subsection), but rather, put all of them in Appendix B. You are allowed up to 12 pages for the graphs produced for your analysis.  Organize the graphs in three main parts, each corresponding to one of the main sections of the technical report.  Also number each of your graphs accordingly to be able to easily refer to them, as necessary, in Sections 1, 2, and 3.  You do not need to repeat graphs in Appendix B. For example, if you want to refer to a graph under the ARIMA section, which was already done in the section dedicated to exponential smoothing, you are encouraged to instead use the figure number of that specific graph rather than repeating the graph again.

2.5. Organizing your technical report. The report must be organized as follows:

1. Exponential smoothing (maximal length: 4 pages; total marks: 40) Marks to be attributed based on how well you articulate the following aspects:

Describe data preparation (and its effects) prior to the implementation of exponential smoothing methods.

Describe preliminary analysis undertaken (and conclusions drawn) prior to the implementation of exponential smoothing methods.

Give details of how exponential smoothing models were selected for each of the time series, and how effective these methods are at forecasting. Clarity and quality of presentation.

Functionality of python codes.

Quality and suitability of illustrative or forecast result graphs.